Institutional variants for lymph node counts after pancreatic resections

Background Lymph node (LN) counts from pancreatectomy are postulated as quality metric for surgical therapy of pancreatic malignancy. Methods Prospectively collected data from a single surgeon's pancreatectomy experience were analyzed for predictors of LN counts. Results Of 315 consecutive patients (54% female, median age: 65, range 18–88), 239 had a proven cancer diagnosis (76%). Operations included pancreatoduodenectomy (69%), distal pancreatectomy (26%), total pancreatectomy (1%) and others (4%). Patients were treated in 4 different tertiary cancer center settings (Institution A: 11%; B: 46%; C: 27%; D: 16%) with consistent regional dissection standards. Mean total LN counts differed between institutions for malignancies (A: 18, B: 13, C: 26, D: 26, p < 0.0001) and benign diseases (p = 0.003). At least 15 LNs were reported in 63% of cancer patients (institution range: 34–92%, p < 0.0001). Conclusions Pathologic processing should be standardized if LN numbers are to be adopted as quality metric for pancreatic cancer resections.